切削刀具在數控機床的加工過程中起著重要作用，繼而影響工件的質量、生產效率、機器穩定性和耐用性。因此在自動化生產系統中應用了刀具磨損監測以提高產品質量。然而之前用於測量的大多數傳感器價格昂貴，且在不同的操作條件下，其結果通常不一致。傳感器設置的位置也限制了處理方法。
因此本研究在CNC機床上安裝了三種類型的傳感器，包括麥克風，加速度計和電流傳感器。首先從訊號的時域、頻域和時頻域中選擇重要的特徵指標，並採用迴歸方法和人工類神經網絡建立刀具磨損模型，最後利用傳感器融合方法建立了刀具磨損預測方法。實驗發現三次多項式模型和人工神經網絡模型均具有良好的預測效果。本研究也比較了使用不同傳感器組合的性能，且將未在訓練資料中、不同銑削參數產生的資料發送到人工神經網絡模型進行驗證。另外當切削信號受到外部噪聲影響時，經過濾波後由三個傳感器融合建立的模型，以及由振動和聲音信號建立的模型具有可接受的預測結果。與單個傳感器模型相比，基於傳感器融合的模型具有更高的通用性和適應性，並可以很好地預測刀具的磨損。

The CNC machine is the main equipment in the modern manufacturing industry. The cutting tool plays an important role in the machining process, which in turn affects the quality of the workpiece, production efficiency, machine stability, and durability. Therefore, the tool wear monitoring was applied in terms of improving product quality, especially in automated production systems. However, under different operating conditions, most of sensors used before were expensive and the results were often inconsistent. The location of the setup also limits the processing method.
Therefore, this study applied three types of sensors on the CNC machine, such as microphone, accelerometer and current transducer. A tool wear prediction method we developed by using the sensor-fusion method. The important feature indexes were selected from the time domain, frequency domain, and time-frequency domain, and regression method and artificial neural network were used to build a tool wear model. It was found that the cubic polynomial model and the artificial neural network model both had good prediction results. The performances by using different combinations of sensors were also compared. The new parameters of milling which are not used for building the model were also sent to the artificial neural network model for verification. The model built by three sensor-fusion and the model made by vibration and sound signals had acceptable prediction results when the cutting signal was subject to the external noise. As a result, the model based on sensor-fusion has higher versatility and adaptability than a single sensor model, and it could make the well prediction on tool wear.

[1]Kious, M., et al. (2010). "Detection process approach of tool wear in high speed milling." Measurement 43(10): 1439-1446.
[2]Liu, C., et al. (2015). "Incremental learning for online tool condition monitoring using Ellipsoid ARTMAP network model." Applied Soft Computing 35: 186-198.
[3]Karandikar, J., et al. (2015). "Tool wear monitoring using naive Bayes classifiers." The International Journal of Advanced Manufacturing Technology 77(9-12): 1613-1626.
[4]吳鴻材, et al. (2017). "基於應變感測的間接式刀具狀態監控." 機械工業雜誌(413): 71-78.
[5]Schwenzer, M., et al. (2019). Machine Learning for Tool Wear Classification in Milling Based on Force and Current Sensors. IOP Conference Series: Materials Science and Engineering, IOP Publishing.
[6]Zhao, R., et al. (2017). "Machine health monitoring using local feature-based gated recurrent unit networks." IEEE Transactions on Industrial Electronics 65(2): 1539-1548.
[7]Mali, R., et al. (2017). "Real time tool wear condition monitoring in hard turning of Inconel 718 using sensor fusion system." Materials Today: Proceedings 4(8): 8605-8612.
[8]Liu, C.-L., et al. (2018). "Time series classification with multivariate convolutional neural network." IEEE Transactions on Industrial Electronics 66(6): 4788-4797.
[9]Ambhore, N., et al. (2015). "Tool condition monitoring system: A review." Materials Today: Proceedings 2(4-5): 3419-3428.
[10]Bhuiyan, M. and I. Choudhury (2014). "13.22—Review of sensor applications in tool condition monitoring in machining." Comprehensive Materials Processing 13: 539-569.
[11]Li, A., et al. (2017). "Effect of cutting speed on chip formation and wear mechanisms of coated carbide tools when ultra-high-speed face milling titanium alloy Ti-6Al-4V." Advances in Mechanical Engineering 9(7): 1687814017713704.
[12]Li, B. (2012). "A review of tool wear estimation using theoretical analysis and numerical simulation technologies." International Journal of Refractory Metals and Hard Materials 35: 143-151.
[13]Chen, H. (2010). Investigation of the Methods for Tool Wear On-Line Monitoring during the Cutting Process. International Conference on Computer and Computing Technologies in Agriculture, Springer.
[14]Siddhpura, A. and R. Paurobally (2013). "A review of flank wear prediction methods for tool condition monitoring in a turning process." The International Journal of Advanced Manufacturing Technology 65(1-4): 371-393.
[15]Ghosh, N., et al. (2007). "Estimation of tool wear during CNC milling using neural network-based sensor fusion." Mechanical Systems and Signal Processing 21(1): 466-479.
[16]Palanisamy, P., et al. (2008). "Prediction of tool wear using regression and ANN models in end-milling operation." The International Journal of Advanced Manufacturing Technology 37(1-2): 29-41.
[17]Nouri, M., et al. (2015). "Real-time tool wear monitoring in milling using a cutting condition independent method." International Journal of Machine Tools and Manufacture 89: 1-13.
[18]Dimla Sr, D. and P. Lister (2000). "On-line metal cutting tool condition monitoring.: I: force and vibration analyses." International Journal of Machine Tools and Manufacture 40(5): 739-768.
[19]Li, X., et al. (2004). "Fuzzy estimation of feed-cutting force from current measurement-a case study on intelligent tool wear condition monitoring." IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews) 34(4): 506-512.
[20]Zhong, Z., et al. (2013). Correlation analysis of cutting force and acoustic emission signals for tool condition monitoring. 2013 9th Asian Control Conference (ASCC), IEEE.
[21]Rmili, W., et al. (2008). Tool wear monitoring in turning process using vibration measurement. Proc. 13th Int. Congress on Sound and Vibration.
[22]Chen, X. and B. Li (2007). "Acoustic emission method for tool condition monitoring based on wavelet analysis." The International Journal of Advanced Manufacturing Technology 33(9-10): 968-976.
[23]Pai, P. S. and P. R. Rao (2002). "Acoustic emission analysis for tool wear monitoring in face milling." International Journal of Production Research 40(5): 1081-1093.
[24]Xiqing, M. and X. Chuangwen (2009). Tool wear monitoring of acoustic emission signals from milling processes. 2009 First international workshop on education technology and computer science, IEEE.
[25]Jemielniak, K. and J. Kossakowska (2010). "Tool wear monitoring based on wavelet transform of raw acoustic emission signal." Advances in Manufacturing Science and Technology 34(3): 5-16.
[26]Doukas, C., et al. (2013). "On the estimation of tool-wear for milling operations based on multi-sensorial data." Procedia Cirp 8: 415-420.
[27]Sevilla-Camacho, P., et al. (2011). "Tool breakage detection in CNC high-speed milling based in feed-motor current signals." The International Journal of Advanced Manufacturing Technology 53(9-12): 1141-1148.
[28]Li, X., et al. (2000). "Real-time tool condition monitoring using wavelet transforms and fuzzy techniques." IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews) 30(3): 352-357.
[29]Neef, B., et al. (2018). Tool wear and surface quality monitoring using high frequency cnc machine tool current signature. 2018 IEEE 16th International Conference on Industrial Informatics (INDIN), IEEE.
[30]Kothuru, A., et al. (2018). "Application of audible sound signals for tool wear monitoring using machine learning techniques in end milling." The International Journal of Advanced Manufacturing Technology 95(9-12): 3797-3808.
[31]Seemuang, N., et al. (2016). "Using spindle noise to monitor tool wear in a turning process." The International Journal of Advanced Manufacturing Technology 86(9-12): 2781-2790.
[32]Liu, M.-K., et al. (2019). "Tool wear monitoring and prediction based on sound signal." The International Journal of Advanced Manufacturing Technology 103(9-12): 3361-3373.
[33]Chen, B., et al. (2011). "Reliability estimation for cutting tools based on logistic regression model using vibration signals." Mechanical Systems and Signal Processing 25(7): 2526-2537.
[34]Aliustaoglu, C., et al. (2009). "Tool wear condition monitoring using a sensor fusion model based on fuzzy inference system." Mechanical Systems and Signal Processing 23(2): 539-546.
[35]Asiltürk, I. and M. Çunkaş (2011). "Modeling and prediction of surface roughness in turning operations using artificial neural network and multiple regression method." Expert systems with applications 38(5): 5826-5832.
[36]Özel, T. and Y. Karpat (2005). "Predictive modeling of surface roughness and tool wear in hard turning using regression and neural networks." International Journal of Machine Tools and Manufacture 45(4-5): 467-479.
[37]Kaya, B., et al. (2011). "Force-torque based on-line tool wear estimation system for CNC milling of Inconel 718 using neural networks." Advances in Engineering Software 42(3): 76-84.
[38]Siddhpura, A. and R. Paurobally (2013). "A review of flank wear prediction methods for tool condition monitoring in a turning process." The International Journal of Advanced Manufacturing Technology 65(1-4): 371-393.
[39]Zhang, X., et al. (2018). "A multi-sensor based online tool condition monitoring system for milling process." Procedia Cirp 72: 1136-1141.
[40]Aliustaoglu, C., et al. (2009). "Tool wear condition monitoring using a sensor fusion model based on fuzzy inference system." Mechanical Systems and Signal Processing 23(2): 539-546.
[41]Kandilli, I., et al. (2007). Online monitoring of tool wear in drilling and milling by multi-sensor neural network fusion. 2007 International Conference on Mechatronics and Automation, IEEE.
[42]Schwenzer, M., et al. (2019). Machine Learning for Tool Wear Classification in Milling Based on Force and Current Sensors. IOP Conference Series: Materials Science and Engineering, IOP Publishing.
[43]Zhang, K.-f., et al. (2015). "A method for tool condition monitoring based on sensor fusion." Journal of Intelligent Manufacturing 26(5): 1011-1026.
[44]Zhou, Y. and W. Xue (2018). "A multisensor fusion method for tool condition monitoring in milling." Sensors 18(11): 3866.
[45]Wang, J., et al. (2017). "Multisensory fusion based virtual tool wear sensing for ubiquitous manufacturing." Robotics and Computer-Integrated Manufacturing 45: 47-58.
[46]Wilamowski, B. M. and H. Yu (2010). "Improved computation for Levenberg–Marquardt training." IEEE transactions on neural networks 21(6): 930-937.